Calorimetric method for determining the thermochemical energy storage capacities of redox metal oxides

Highlights

• Method for accurately measuring enthalpy of metal oxides for thermochemistry.

• Catalysts enable accurate solution calorimetry data for metal oxides in strong acids.

• Facilitates accurate computation of energy storage potentials of novel metal oxides.

• Distinguishes energy stored as sensible heat from that as reaction energy.

• Results shown for novel magnesium-manganese oxide energy storage material.

Abstract

Transition metal oxide compounds have been demonstrated to be promising candidates for thermochemical processes, particularly energy storage. A calorimetric method for measuring enthalpy of these materials is developed in this work. A combination of drop calorimetry and acid-solution calorimetry is used to measure the total enthalpy and standard enthalpy of formation of these materials for compounds that form at high temperatures (≥ 1000 °C) after undergoing thermal reduction. These measurements are used to compute the energy storage potential of these materials for a specified set of redox cycle operating temperatures. The construction and calibration of appropriate calorimeter devices are described. A novel approach to performing acid-solution calorimetry is introduced by including tin (II) chloride as a reducing agent for enhancing the dissolution rate of these compounds sufficiently so that acid-solution calorimetry can be implemented accurately. The general procedure is presented and applied to three variations of magnesium-manganese oxide materials for a redox cycle operating under an oxygen partial pressure of 0.2 atm between 1000 and 1500 °C. Using this method, the total energies stored by magnesium-manganese oxides of molar ratios Mn/Mg of 2/1, 1/1, and 2/3 were found to be 924 ± 56.7, 1029 ± 57.0, and 1070 ± 64.2 kJ kg⁻¹, respectively.

Introduction

As renewable power generation increases worldwide, the need for advanced energy storage methods continues to grow. Solar energy in particular requires the development of efficient storage systems in order to overcome its intermittent nature and meet daily energy demands [1]. One such method of interest is thermochemical energy storage (TCES). These systems are highly attractive: they provide higher energy densities and improved thermal efficiencies through higher operating temperatures relative to existing sensible or latent energy storage systems, and can be implemented for peak-production storage across a wide variety of power generation methods [2]⁻ [3].

The current focus in this field is the development of materials that are reactively stable and exhibit high energy storage capacities at high temperatures [3]. Transition metal oxides have been shown to be good candidates for TCES via reduction/oxidation reactions at high temperatures. Dizaji et al. [2] and Wu et al. [3] performed comprehensive reviews of the current literature on both pure and mixed metal oxide redox TCES systems under study, finding that in general, these systems show promise for their high operating temperatures (≥ 700 °C), long energy storage durations, high storage capacities due to high reaction enthalpies, and the ability to be oxidized in air, avoiding complex gas storage systems. Of the variety of metal oxides considered, manganese oxides appear to be an attractive candidate for their low toxicity and low costs relative to other systems, such as Co₃O₄ and BaO₂ [3]. Studies on the pure Mn₂O₃/Mn₃O₄ redox couple have all identified very poor oxidation kinetics as the main drawback of pure manganese oxides [[4], [5], [6], [7], [8], [9]]. Subsequent studies have attempted to enhance oxidation kinetics by mixing manganese oxides with a variety of other transition metal oxides, to varying degrees of success [1,2,5,[10], [11], [12], [13], [14], [15]].

To improve the viability of manganese-oxide-based TCES, Randhir et al. [16] mixed magnesium oxide and manganese oxide to form a magnesium-manganese oxide redox material that exhibits the following reversible reaction at high temperatures,

$$\text{M}{\text{g}}_{\text{x}}\text{M}\text{n}{{\text{O}}_{1+\text{x}+\text{y}}}_1\leftrightarrow \text{M}{\text{g}}_{\text{x}}\text{M}\text{n}{\text{O}}_{1+\text{x}+\text{\hspace{0.17em}y}2}+\frac{{\text{y}}_1-{\text{y}}_2}{2}{\text{O}}_2.$$

Here, x is the ratio of Mg/Mn, and y₁ and y₂ are the excess oxygen contained within the material lattice for oxidized and reduced materials, respectively. It should be noted that Eq. 1 is a simplified version of the chemical reaction, and a comprehensive discussion on the reaction evolution can be found in Randhir et al. [16]. In summary, the material functions as a storage medium by decomposing from a spinel structure to a monoxide structure as the temperature is increased from 1000 to 1500 °C, releasing oxygen gas in the process. The work of Randhir et al. with magnesium-manganese oxides demonstrated both reactive stability and enhanced speed of oxidation while operating between 1000 and 1500 °C for materials with Mn/Mg molar ratios of 2/1, 1/1, and 2/3, known hereon as 2/1 MM, 1/1 MM, and 2/3 MM, respectively [16].

The purpose of this study is to measure the enthalpy of magnesium-manganese oxides in order to compute their energy storage capacities for a specified redox cycle. To do so, a combination of drop calorimetry and acid-solution calorimetry are utilized to measure the total enthalpies and the standard enthalpies of formation of compounds that form at temperatures of interest for thermochemical energy storage applications. Both techniques have been successfully utilized in the literature. Drop calorimetry has been employed quite commonly for measuring enthalpy of metal oxides at high temperatures. For example, Frederick et al. [17] measured the enthalpy of uranium dioxide up to 1227 °C. Popa et al. [18] measured the thermodynamic properties of monazite from 177 to 1227 °C. Takahashi et al. [19] measured the enthalpy and specific heat of gadolinia-doped uranium dioxide between 127–727 °C. Ritchet et al. [20] measured the thermodynamic properties of several silicon dioxide phases between 727–1527 °C. Acid-solution calorimetry has also been shown to be a successful technique. Brink and Holley [21] used this technique to calculate the standard enthalpy of formation of strontium (II) oxide by dissolving it and pure strontium metal in aqueous 1 M hydrochloric acid solution (HCl). Matskevich et. al. [22] successfully calculated the standard enthalpy of formation of SrCeO₃ perovskite doped with lutetium (III) oxide by dissolving samples in a solution of 1 M HCl. Tsvetkov et al. [23] calculated the standard enthalpy of formation of double perovskites GdBaCo_2−xM_xO_6-δ (M = Fe, Mn; x = 0, 0.2) by dissolving constituent metal-oxides in 200 mL of 4 M HCl solution.

Fig. 1 illustrates how these two techniques are applied to determine the energy storage capacities of magnesium-manganese oxides. When operating at high temperatures and a constant oxygen partial pressure, the energy stored by magnesium-manganese oxides is separated into two components: that stored as chemical reaction energy, and that stored as sensible and latent energy. Cooling the material under an inert gas allows for the latent and sensible heat to be released while preventing heat release due to oxidation. By preventing oxidation, the enthalpy at temperatures of interest can be measured using drop calorimetry for both the reduced and oxidized material states. This allows us to isolate the energy stored as sensible and latent heat as the temperature is increased. The energy released due to oxidation can then be calculated as the difference between the standard enthalpies of formation of the reduced and oxidized materials, measured using acid-solution calorimetry. The total energy storage capacity for a specified temperature cycle is then computed by,

ΔH_S = (ΔH_{Drop, RED} – ΔH_{Drop, OX}) + ΔH_Rxn

where ΔH_{Drop, RED} is the enthalpy difference between 1500 and 25 °C for reduced materials, ΔH_{Drop, OX} is the enthalpy difference between 1000 and 25 °C for oxidized materials, and ΔH_Rxn is the chemical energy stored during thermal reduction measured using acid-solution calorimetry.

The novelty of our method has two facets. The combined use of drop and acid-solution calorimetry allows the change in enthalpy due to chemical reaction and that due to sensible heat (and phase change if present) to be distinctly isolated. In addition, to obtain accurate measurements using acid-solution calorimetry, the material under study must dissolve quickly within the acid. When the material takes a long time (order of hours) to dissolve, the concurrent heat loss from the system severely reduces the accuracy of the calorimetry data. Many metal oxides, in particular, dissolve very slowly in strong acid solutions. However, by adding the right catalyst to the acid, the rate of dissolution of the material of interest can be enhanced significantly, enabling accurate solution calorimetry measurements to be taken. To facilitate this, an acid-solution calorimetry method was developed by adding tin (II) chloride (SnCl₂) to aqueous hydrochloric acid solution to enhance the rate of dissolution of magnesium-manganese oxides sufficiently for calorimetry.

In the proposed method, we’ve combined catalytic dissolution enhancement for acid solution calorimetry with drop calorimetry to develop a methodology to study reactive metal oxides. The method delineates the energy stored due to chemical reaction and that stored due to sensible heat change (also phase change if present). We will show that this method is very useful for measuring the energy storage capacity for reactive metal oxides. Further, we believe this method offers an accurate, simple, and low cost alternative to differential scanning calorimetry (DSC) commonly used in the study of high temperature thermochemistry. DSC is the most common technique employed for studying the energy storage capacity of TCES redox metal oxides, as the vast majority of studies compiled by Dizaji et al. [2] and Wu et al. [3] utilized DSC in their analyses. However, it has some significant drawbacks when applied to high temperature TCES metal oxides. Jacob et al. [24] examined the accuracy and reproducibility of specific heat measurements using DSC at high temperatures (> 700 °C), finding that results at temperatures ≥ 800 °C were highly variable with variations in sample mass, mass of the calibration standard, heating rate, and flow rate of the purge gas. Saeed et al. [25] studied the uncertainties associated with the enthalpy of melting and transition temperature measurements of phase change materials taken using DSC, finding that using high heating rates and/or large sample masses resulted in shifts towards results of higher magnitude. The low thermal conductivities and high energy storage densities of phase change materials create a noticeable temperature gradient within the sample, resulting in a higher temperature measured by the DSC thermocouple than the actual average sample temperature, thus creating a biased measurement. Small sample mass and heating rate reduce the thermal gradient bias, but widen the range of uncertainty due to a low observable heat flux [25]. These two studies hold relevance for TCES metal oxides at high temperatures (≥ 800 °C). The first implies that the low accuracy and reproducibility above 800 °C will greatly increase the uncertainty in enthalpy measurements collected using DSC at temperatures of interest (1000–1500 °C) for TCES purposes. In addition, redox metal oxides have high energy storage densities and low thermal conductivities (when subjected to high temperatures). Therefore, the temperature bias caused by the required fast heating rate and inherent temperature gradient make it very difficult to accurately measure reaction enthalpies of redox metal oxides at high temperature using DSC. We initially attempted magnesium-manganese oxide energy storage capacity measurements using DSC but were not satisfied with the accuracy due to the measurement issues described above.

The method we propose utilizes a simple, accurate, and inexpensive means of measuring standard enthalpy of formation for compounds that form at high temperature (≥ 1000 °C). The ability to analyze several grams (compared with mg samples for DSC) of sample at once reduces the potential for material non-uniformity error. Utilization of a large tube furnace ensures precise temperature control of the sample during the heating step, as well as significant cost savings over a thermogravimetric/DSC analyzer. Additionally, this method facilitates the computation of enthalpy differences between the standard state and the temperature of interest for redox metal oxides. These measurements can then be used to compute the total energy stored between two temperatures of interest for a thermochemical process, providing a clear distinction between the contribution from chemical reaction and that from sensible and latent heat change. Overall, we believe this approach towards measuring enthalpy change of unstudied metal oxide materials can be a very useful tool for thermochemistry research involving metal oxides.